Contrastive News and Social Media Linking using BERT for Articles and Tweets across Dual Platforms
Jan A. PiotrowskiMarek WachnickiMateusz PerlikJakub PodolakGrzegorz RuckiMichał BrzozowskiPaweł OlejnikJulian KozłowskiTomasz NocońJakub KoziełStanisław GizińskiPiotr Sankowski
0
Citation
0
Reference
10
Related Paper
Abstract:
X (formerly Twitter) has evolved into a contemporary agora, offering a platform for individuals to express opinions and viewpoints on current events. The majority of the topics discussed on Twitter are directly related to ongoing events, making it an important source for monitoring public discourse. However, linking tweets to specific news presents a significant challenge due to their concise and informal nature. Previous approaches, including topic models, graph-based models, and supervised classifiers, have fallen short in effectively capturing the unique characteristics of tweets and articles. Inspired by the success of the CLIP model in computer vision, which employs contrastive learning to model similarities between images and captions, this paper introduces a contrastive learning approach for training a representation space where linked articles and tweets exhibit proximity. We present our contrastive learning approach, CATBERT (Contrastive Articles Tweets BERT), leveraging pre-trained BERT models. The model is trained and tested on a dataset containing manually labeled English and Polish tweets and articles related to the Russian-Ukrainian war. We evaluate CATBERT's performance against traditional approaches like LDA, and the novel method based on OpenAI embeddings, which has not been previously applied to this task. Our findings indicate that CATBERT demonstrates superior performance in associating tweets with relevant news articles. Furthermore, we demonstrate the performance of the models when applied to finding the main topic -- represented by an article -- of the whole cascade of tweets. In this new task, we report the performance of the different models in dependence on the cascade size.Keywords:
Viewpoints
Representation
With the rapid development of microblog, the research of topic detection has been paid more attention, which has begun to transfer from the traditional news media to microblog. However, compared with the traditional fields where topic detection applied, the microblogging data is written informally and the structure of that is not rigorous, which will bring great difficulties. In the process of topic detection, there will be lots of noise information extracted as the topics if using the traditional methods. Therefore, based on the Latent Dirichlet Allocation (LDA) model, the Topic-Coreterms Latent Dirichlet Allocation (TC-LDA) model for topic detection is proposed. With considering the microblogging features, the model increases a layer of background model to reduce the noise in microblogging data. The experiment shows that through the method of extending, the model can obtain good results and improve the generalization performance in the same environment.
Microblogging
Cite
Citations (10)
This study focuses on how products are viewed by their users. In order to capture users' viewpoints, we focused on how the virtual 3D model of a product was examined (rotated/translated/moved by the user). We attempt to understand these viewing mechanisms by focusing on the dynamics of shifting viewpoints and verbal features during the generation of product impressions. The notion of dynamics of shifting viewpoints in this study refers to changes between 'still viewpoints' and 'moving viewpoints'. If, in a given period when product impressions are generated, a viewpoint does not change, it is defined as a still viewpoint, whereas a moving viewpoint refers to shifts in the ways in which the products are viewed during a given period. The notion of verbal features involves the quantitative characteristics of verbalisations. The findings from an experimental observation show that the dynamics of shifting viewpoints, verbal features, and user's preferences are related.
Viewpoints
Dynamics
Cite
Citations (2)
Previous commentary systems generate commentaries only from the viewpoint of a commentator. However there are various viewpoints for comments, and such different viewpoints invoke various comments. The amount of information about a situation may differ between the viewpoints, and the understandings of the situation may also differ between them. In this paper, we propose a method to generate commentaries automatically so that users can easily understand situations by taking into account the different understandings of the situations between viewpoints. Our method is composed of two parts. The first is generation of comment candidates about the current situation, unexpected actions, intentions of players by using a game tree. The second is comment selection which chooses comments related to the prior one so that listeners can compare the situations from different viewpoints. Based on our approach, we implemented an experimental system that generates commentaries on mahjong games. We discuss the output of the system.
Viewpoints
Tree (set theory)
Cite
Citations (0)
This article explores the cognitive processes by which Year 7 students (aged 11 to 13 years) use personal viewpoints to interpret and resolve life-issues. A Christian school located in the North West of Sydney wanted to find out if the proportionally higher occurrence of discipline issues in newly enrolled Year 7 students could be explained by those students having a more diverse range of personal viewpoints than Year 7 students who had been enrolled for a longer time. To address this question, we conducted a survey of Year 7 students at the school in early 2015. The survey consisted of five open-ended questions related to a hypothetical school bullying situation. The students’ answers were coded according to the various types of personal viewpoints they used to form an interpretation of the scenario event and to resolve the issue. The students’ responses reflected 18 personal viewpoints across the five questions. Their answers varied in their use of complex personal viewpoints – that is, answers consisting of two or more simple personal viewpoints. A pattern was discerned from the varying proportions of complex viewpoints each question elicited, some questions drawing a high proportion of complex viewpoints. The differing proportions of complex personal viewpoints used in answering each of the five questions suggested that a higher proportion of complex personal viewpoints indicated the occurrence of enhanced complex thinking. A further finding was that students used a higher proportion of complex personal viewpoints when answering questions that challenged them to take a third-person perspective on the protagonists in the narrative.
Viewpoints
Cite
Citations (5)
To research the differences of the viewpoints on women of Western and China before 18th century,by making comparison of viewpoints on women between in earlier western countries and in earlier China from three aspects,viz.religionary phenomenons,philosophic thinking ways,and literary descriptions,to aim at finding out respective characteristics of early viewpoints on women in China and in western countries.The study attempts to present cultural identity and distinguish about viewpoints on women between the East and the West.Also it discusses social effects and history effects of respective viewpoints on women.Because of differences on social structure and cultural background,the earlier viewpoints on women between Western and China are with obvious distinctions,also these distinctions have very important referential meanings to the research on the forming and developing of the viewpoints on women in the present society.
Viewpoints
Cite
Citations (0)
In order to provide a notification system for the blind awaiting a bus, a classification of viewpoints is necessary. In a previous study, we classified the viewpoints in non-congested traffic using road features extraction. However, the viewpoints in congested traffic have not yet been discussed; hence this paper proposes a viewpoints classification using car distribution on the road in congested traffic. The study aimed to classify the viewpoints to two classes as "Good Viewpoints" and "Bad Viewpoints" with 100 images for each class. The experimental results showed a 76% accuracy of the proposed method.
Viewpoints
Cite
Citations (1)
We present a field study aimed at analysing the use of viewpoints in co-design meetings. A viewpoint is a representation characterised by a certain combination of constraints. Three types of viewpoints are distinguished: prescribed viewpoint, discipline-specific viewpoint and integrated viewpoint. The contribution of our work consists in characterising the viewpoints of various stakeholders involved in co-design (design office disciplines, and production and maintenance disciplines), the dynamics of viewpoints confrontation and the cooperative modes that enable these different viewpoints to be integrated.
Viewpoints
Representation
Cite
Citations (0)
Five classical viewpoints about wisdom will be commented briefly one by one in this paper,which contains Piaget's viewpoints of intelligence,Erikson's viewpoints of Wisdom,viewpoints of Wisdom created by the Berlin Model of Wisdom,viewpoints of Wisdom created by neo-Piagetism,and Sternberg's viewpoints of Wisdom.
Viewpoints
Cite
Citations (1)
A variety of viewpoints were voiced at the workshop either in oral presentations or in the ensuing discussions. These viewpoints have been collected and organised in six broad areas. Workshop participants have contributed by reviewing and commenting on the present collection of viewpoints, which is provided for convenience. More detailed and precise information is provided in the summary of the workshop and the contributed papers.
Viewpoints
Cite
Citations (0)
The nationally-recognized Susquehanna
Chorale will delight audiences of all
ages with a diverse mix of classic and
contemporary pieces. The ChoraleAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂA¢AÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂs
performances have been described
as AÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂA¢AÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂemotionally unfiltered, honest
music making, successful in their
aim to make the audience feel,
to be moved, to be part of the
performance - and all this while
working at an extremely high
musical level.AÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂA¢AÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂAÂA Experience choral
singing that will take you to new
heights!
Cite
Citations (0)